Keenan Down

Profile

Project Title: Using information theory to diagnose Alzheimer's disease and predict 
disease progression

Summary: 

I am investigating partial information decomposition (PID) as a tool to understand the progression of Alzheimer's disease. In addition, I hope to develop algebraic tools to characterise information 'flow' and exchange so that these might also be applied to brain activity.

Partial information decomposition allows us to characterise interactions between several variables as redundant, unique, or synergistic. The standard example is human vision; much information is provided by both eyes redundantly (such as colour), or uniquely (at the edge of your vision). When using both eyes, one gets information about depth that was not accessible with one eye alone. This 'synergistic' information component reflects an integration of available information.

As leading theories of consciousness (such as IIT) depend on integration, the synergy of interacting variables can be taken to reflect consciousness and connectedness. With the emerging view that Alzheimer's disease is a disconnection syndrome and hence a disease of consciousness, PID and the more recently developed integrated information decomposition (Phi ID) open new avenues for exploring and understanding the onset of Alzheimer's disease.

Supervisor:

• Dr Daniel Bor

Research